Summary of presentations
The papers on atmospheric observations focused on four primary areas namely requirements, programmes with existing in situ observation networks, future observing systems, and a regional framework for observations. A brief summary of each paper is given below as well as a summary of the ensuing discussion.
Guidelines and principles of climate monitoring
Prior to implementing changes to existing systems or introducing new observing systems standard practice should include an assessment of the effects on our ability to monitor climate variations and changes. Overlapping measurements of both the old and new observing systems should be standard practice for critical climate variables.
Calibration, validation, knowledge of instrument, station and/or platform history is essential for data interpretation and use. Changes in instrument sampling time, local environmental conditions, and any other factors pertinent to the interpretation of the observations and measurements should be recorded as a mandatory part of the observing routine and be archived with original data. The algorithms used to process observations must be well documented. Documentation of changes and improvements in the algorithms should be carried along with the data throughout the data archiving process.
The capability must be established to routinely assess the quality and homogeneity of the historical database for monitoring climate variations and change including long-term high resolution data capable of resolving important extreme environmental events. Environmental assessments that require knowledge of climate variations and change should be well integrated into a Global Observing Systems strategy.
Observations with a long uninterrupted record should be maintained. Every effort should be applied to protect the data sets that have provided long-term homogeneous observations. Long-term may be a century or more. Each element of the observation system should develop a list of prioritized sites or observations based on their contribution to long-term environmental monitoring.
Data poor regions, variables, regions sensitive to change, and key measurements with inadequate temporal resolution should be given the highest priority in the design and implementation of new environmental observing systems.
Network designers, operators, and instrument engineers must be provided climate monitoring requirements at the outset of network design. This is particularly important because most observing systems have been designed for purposes other than long-term monitoring. Instruments must have adequate accuracy with biases small enough to resolve climate variations and changes of primary interest.
Much of the development of new observation capabilities and much of the evidence supporting the value of these observations stem from research-oriented needs or programmes. Stable, long-term commitments to these observations, a clear transition plan from research to operations, are two requirements in the development of adequate climate monitoring capabilities.
Data management systems that facilitate access, use, and interpretation are essential. Freedom of access, low cost, mechanisms which facilitate use (directories, catalogues, browse capabilities, availability of metadata on station histories, algorithm accessibility and documentation, etc.) and quality control should guide data management. International cooperation is critical for successful management of data used to monitor long-term climate change and variability.
IPCC requirements
"Warmer temperatures will lead to a more vigorous hydrological cycle; ..."
The following questions are raised by this conclusion of the IPCC (1995) report.
Q.1 Because the statement appears to be mostly based upon "medium" to "low" confidence data for the time period of 1950s to early 80s (when the global temperature remained largely flat) should we not be extending the database into the 1990s when the warming seems to have really picked up?
Q.2 Which of the hydrological cycle components will show the most vigour and where? Should we not be strengthening the observing system for those parameters and in those regions with a higher priority?
Q.3 What changes in advection and convection are implied by enhanced spatial gradients predicted in the warmer world? Should we not be monitoring those changes also? A second monitoring issue which is raised by the latest IPCC Assessment is that although the future climates over the continents appear to be mostly dependent on the aerosol concentrations over these areas, there is currently no reliable observation system to monitor the distribution of optical characteristics of aerosols either for land or ocean.
World Weather Watch in situ observations
Most likely any global observing network of the atmosphere will include the WWW either directly or indirectly. It should be noted that a focus on that small portion of the WWW that deals with the in situ aspects of the surface-based Global Observation System (GOS) will present a limited capability. The true capability is better reflected in a full understanding of the Basic Systems of the WWW. The surface-based GOS consists of six Regional Basic Synoptic Networks (RBSN) composed of about 4000 surface and 900 upper-air stations. The GOS also includes a further 6000 land surface observing stations, 7300 Volunteer Observing Ships, 600 data buoys, a large portion of the commercial wide body airliners that provide up to 40,000 observations per day, and many other observing technologies that primarily are directed at national and high resolution observations such as Radar and Sferics.
The overall trend of the GOS is to more automation and remote sensing. With respect to the surface-based RBSN, the surface observations are increasing and the quality of observations are improving due to increased automation. The upper air network is declining but at a very slow rate. The basic problems are the pending termination of the Omega navigational system and the increase cost of expendable items. With respect to climatological and agricultural meteorological stations, there is a continuing increase at a fairly moderate rate, that is expected to continue. The number of Volunteer Observing Ships (VOS) is slowly decreasing due to the overall decline of the merchant marine service; however the quality and quantity of ship reports is actually slightly increasing due to automation and more efficient communication through satellites. The major growth in the surface in situ system will be in data buoys (drifting and moored) and in aircraft observations, again principally because of automation and more efficient telecommunications. With respect to factors related to improvements and impediments to improving the surfaced-based observing system, there are several, which in some cases pertain to the overall WWW. The major impediments include increased costs to developing countries and the ability to link synoptic type observations to national services and priorities. The major factors related to improvements are as follows:
a) implementation is linked to a requirements process rather than a static set of requirements - the opportunity for technology and scientific upgrades are built in and include complementary research programmes;
b) there are mechanisms for all countries to participate through contributions to the programme, derivation of benefits from the programme, and a structure for assistance and cooperation activity specifically in support of the WWW;
c) implementation and coordination groups are composed of the actual "hands-on" people;
d) exposure of the WWW Programme at the highest level of the UN system (UN General Assembly resolution);
e) a flexible and creative set of funding mechanisms involving both national and regional mechanisms.
Global Atmosphere Watch
The main goal of the WMO Global Atmosphere Watch (GAW) is to coordinate the ground-based long-term monitoring system of the chemical composition of the atmosphere and related parameters such as greenhouse gases, aerosols, pollutants, precipitation chemistry, and ozone. GAW is contributing to the evaluation and the understanding of the influence of the atmospheric composition on the environment as well as to the understanding of the climate variability. GAW is therefore contributing to GCOS. GAW also provides scientific assessments of long-term trends and early warning of changes in composition of the atmosphere.
During the past 4 years, the main effort has been in developing the network coverage and in improving the final scientific quality of the measurements. In this respect, eight new global stations which combine full atmospheric chemistry measurements and scientific activity (six in developing countries supported by the Global Environmental Facility (GEF)) have been implemented bringing the total to 21.
The regional network has also been enhanced, particularly in the southern cone region of South America where 11 new stations for ozone and ultra-violet radiation measurements have been implemented.
Concerning the quality of the data provided by GAW, three Quality Assurance Centres associated with 12 World Calibration Centres have been implemented in order to ensure the scientific quality of the data archived in the six WMO World Data Archiving Centres responsible for providing the data to the users: scientific community, agencies and governments.
For the future, the main priority of the GAW programme is to consolidate the present developments and ensure their long-term continuation:
Reinforcement of the Quality Assurance/ World Calibration Centres system;
Continuation of the present good cooperation with international programmes: International Global Atmospheric Chemistry Project (IGAC), World Climate Research Programme (WCRP), GCOS;
More training, particularly in developing countries that are an essential support to the programme;
Support of scientific assessment of GAW issues.
Aircraft and Future Observing Systems
Global coverage from satellites is crucial for the study of climate change. However, with a few exceptions, the satellite error bars are still too large for some climate applications. The application of Kalman filters to quantify and to extract gradient information from satellite data requires high space/time density of in situ (or other remotely sensed) measurements. This can only realistically occur with the addition of new automated in situ and/or other remotely sensed information sources.
Four key examples of automated observing systems were provided which would help fill crucial gaps in our understanding of the climate system. The first example is the use of many Profiling Automatic Lagrangian Circulation Explorer (PALACE) floats to help deduce upper ocean thermal and density structure. This can be achieved with the help of the satellite altimeter.
The second example is the use of ground-based remote sensing systems for water vapour and wind measurements, along with similar measurements from commercial aircraft, to obtain low-level water vapour flux information. These same automated systems for water vapour can be used with the satellite water vapour measurement systems to significantly improve water vapour flux measurements associated with large scale monsoon active and break periods. These combined systems would also be used to better establish the important upper level moisture fields associated with cirrus cloud formation and greenhouse gas warming issues.
The third example comes from the carbon dioxide assessment community, which is calling for more detailed four-dimensional measurements of carbon dioxide. Here, a method of using package-carrying commercial aircraft as a means for increased chemical measurements has been identified. Also possible with these package-carrying aircraft are wind profiles over the ocean using Doppler lidar systems. These wind measurements, while valuable in their own right, would provide an additional means for calibration and validation of future Doppler lidar systems on micro-satellites.
The fourth example is the establishment of several "super sites" at various surface locations. Such sites would measure surface fluxes, provide remotely sensed profiles of temperature and humidity, be equipped with a cloud radar, and also contain a pulsed lidar. These systems would simultaneously measure radiation, clouds and aerosols and provide a means for the calibration and validation of similar simultaneous field of view satellite measurements with similar instruments. Such a satellite mission is being proposed by Global Energy and Water Cycle Experiment (GEWEX).
All of these examples are systems with very low recurring costs. However, initial capital costs are not insignificant, though still extremely small compared to satellite costs. This requires a new concept of shared satellite/non-satellite observing systems for funding and planning.
North American Atmospheric Observing System
The NAOS programme, currently being implemented by Canada, Mexico and the USA, is a strategy to define the "best mix" of observing systems, reduce observing systems risks and uncertainty, and improve the government decision-making process in defining the future atmospheric observing system.
During a time when the present data set is inadequate to further meteorological services, the maintenance of existing systems is becoming more costly, changes are being mandated, new observing systems are emerging or are in limited use, and the NAOS strategy is being put in place to assist in making the difficult choices ahead. A mix of scientific evaluation of alternative observing architectures, operating system considerations, and systems design is being called upon to assist in making the difficult decisions for implementing alternative observing system strategies.
A NAOS Council has been established with representation from NOAA, the research community, Canada, Mexico, and other US Federal agencies. Three Council sessions have been held on a quarterly basis, a test and evaluation set of hypotheses have been approved. Studies have been initiated and are focused upon improvements to the Upper Air Sounding System. Early efforts will focus upon radiosondes, automated aircraft data, wind information from profilers and an expanding network of Doppler radars (WSR-88D), but eventually all observing system components will be examined.
Results will be presented at annual sessions of the American Meteorological Society and other venues. An Outreach Plan is in preparation and a programme plan is in press. The strategy of implementation includes high level scientific and management participation in the Council who can take direct action with existing computational and human resources and have a management mechanism to report to decision-makers in the US, Canada, and Mexico. No major new system implementation decisions will be taken without adequate scientific and technical supporting information because of the considerable funding implications. The impact on climate measurements will be an explicit consideration.
The Composite Observing System for the North Atlantic (COSNA)
The session was informed of the structural and organizational set-up of the COSNA that has been supported for seven years by several national meteorological services (NMSs) on the basis of a non-governmental agreement. The COSNA Agreement aims at the implementation of the WWW components especially with regard to generating atmospheric observational data needed for numerical weather prediction (NWP) on a real-time operational basis.
Maintaining COSNA as an operational system involves:
A scheme of voluntary contributions in cash and in kind;
A scientific evaluation group (SEG) supported by the major NWP centres;
A small trust fund, fed by regular contributions in cash;
Very small overhead costs.
The operational components of COSNA make use of all available data sources in the COSNA region that are coordinated and integrated in a systematic manner that includes regular performance monitoring and evaluation against data requirements stated by the users. Use is made in particular of the WWW specialized monitoring centres and of input provided by the operators of systems components like ASDAR/AMDAR (aircraft platforms), ASAP (automated shipboard aerological programme), data buoys, voluntary observing ships.
The SEG focuses on studying the relative merits of the COSNA components with a view to optimizing the system design using impact studies and input from various data assimilation schemes. This includes involvement of the major research groups in the NWP area. COSNA provides a unique example of NMSs taking joint responsibility for operational observing systems operated outside national boundaries in a data-sparse area. It was suggested that COSNA might serve as a model for establishing the global observing system modules by creating joint schemes for funding, planning and operation by Members. The use of existing agreements such as that for COSNA was offered as an opportunity to serve the purpose of GCOS provided the special requirements are made known to the Co-ordination Group for the Composite Observing System for the North Atlantic (CGC). In an environment of competition for funding resources, budget cuts and dominating national interests, the COSNA approach seems to offer a viable solution to the start-up problems and future operation of relevant modules of the GCOS.
It was noted that a group of West European NMSs is now in the process of designing a composite observing system that will include COSNA. The new agreement includes joint funding by assessed contributions and, while assigning joint responsibility for implementing and maintaining the relevant operational structures, will dominate national interests.
Plenary Discussion
A variety of factors are driving observation system changes including limitations on resources, emerging new technologies that offer the possibility of reduced costs and improved observations, new requirements based on increased understanding of natural and anthropogenic climate variability and change, and the general recognition of the need to improve past data as well as present and future observations.
Two views emerged with respect to how well observing systems are integrating these factors into present and proposed networks. Scientists trying to better understand climate variability and change, for example as reported to the IPCC, have identified a variety of observing system inadequacies, whereas network managers and users, focusing on the short-term prediction problem, appear to be more satisfied with the current state of affairs. It was pointed out that the adequacy of observations is dependent upon the questions being addressed. For example, with respect to precipitation measurement, current biases can be of minor significance in short-term analyses of rainfall, but they are very significant for long-term monitoring of rain and snowfall. This led to the point that the most serious inadequacies and perhaps the most difficult requirements to achieve are likely to be related to detecting climate change and variability. In this context it was suggested that the Guidelines and Principles for Climate Monitoring were applicable to many other areas, and each of the global observing systems should review the Guidelines during their general requirements-setting process.
Several other issues arose during the discussion related to a variety of data-quality strategies. It was pointed out that re-analysis and data assimilation schemes are currently being used to improve data quality and extend data to observation-sparse areas. The importance of near-real-time data access from data distribution centres is crucial for both improved data quality and assessments. As data sets change it is critical to retain previous versions and make them all readily accessible. Although the number of reporting stations for the WWW over the past few decades has been nearly constant, or even increasing (depending on the network), there has been a redistribution of stations such that gaps in coverage have been emerging, e.g., data-poor areas have fewer stations now compared to previous decades.
The discussion concluded with the question "What is currently broken in the observing system and what needs to be fixed?" The answer seems to depend on the scientific question being asked. For example, addressing the question "Which components of the hydrological cycle are expected to increase in intensity?" might lead to a different set of priorities compared to addressing the question, "Which aspects of the hydrological cycle will have the greatest impact on managed and natural ecosystems?" Nonetheless, if some key elements could be recommended for improved measurements and specific characteristics be developed to achieve these measurements, then the JSTC may be able to consolidate and articulate the global observing system's highest priorities to the stake-holders and network operators.
The need was noted to correlate the in situ observation sites identified through GTOS and the climate observation sites selected by GCOS, to determine the extent of useful overlap and the potential for linkages between sites.
Recommendations
Recommendation A1. It is recommended that the Sponsors of GCOS, GOOS and GTOS adopt the "Guidelines and Principles for Climate Monitoring" within those observation and data programmes that support the global observing systems and the studies of the IPCC. In particular, consideration should be given to:
Studies to assess impacts of the new technology as they affect the climate record preferably prior to implementation. This information should be widely distributed;
Wide distribution of periodic information regarding calibration and metadata and processing of data;
Periodic information on data quality monitoring and data assessment including random errors and long-term systematic biases from the appropriate monitoring centre;
Establishment of Reference Networks similar to those established for surface and upper air within the WWW for GCOS.
Recommendation A2. The Sponsors should consider a mechanism for the scientific advisory bodies to consult with the IPCC regarding data requirements for future assessments.
Recommendation A3. The WMO should make available updates of baseline data sets such as COADS on a regular basis.
Recommendation A4. The WMO should consider ways of expanding and improving the aircraft observing scheme to include additional aircraft as well as new parameters such as atmosphere moisture and carbon dioxide.
Recommendation A5. In implementing their plans the global observing systems should take advantage of observations and products being derived through new regional and platform observing management activities such as NAOS and COSNA as ways of expanding global observing capabilities (Recommendation A5).
Recommendation A6. The scientific and steering advisory groups of the global observing systems should continue to conduct scientific evaluation through impact and sensitivity studies so as to influence the design and evolution of the observing systems in cooperation with existing scientific committees.
Recommendation A7. The Sponsors should continue to support inter-programme data management.
Recommendation A8. The Sponsor's policies should ensure full and open access of all data required for the global observing systems.
Recommendation A9. The GCOS JSTC should consider taking the lead responsibility for the cryospheric and hydrologic component of climate but that the subject of climate impacts on terrestrial systems should be carried out jointly with GTOS.
Recommendation A10. The GCOS JSTC should facilitate the enhancement and development of a global database on aerosol characteristics appropriate to quantify regional and global climate forcing.
Recommendation A11. It is recommended that an analysis be carried out to evaluate the extent of the overlap and possible linkages between the in situ ecological sites identified by GTOS and the base-line observation system definedby GCOS.
